Resident macrophages in adipose tissue account for much of the cytokine expression with obesity. We described the expression of thrombospondin (TSP1), a pleiotropic adipokine, in insulin resistant subjects. Through its anti-angiogenic properties and its ability to activate TGF2, TSP1 may be an important part of the adipocyte/macrophage interaction that results in simultaneous impairment of angiogenesis and increased TGF- 2-mediated fibrosis and inflammation. We recently demonstrated that macrophages also infiltrate skeletal muscle with obesity that is strongly correlated with insulin resistance. Furthermore, the presence of macrophages in co-culture with muscle cells, in the presence of palmitic acid to mimic an obese environment, results in a synergistic increase in the expression and secretion of inflammatory cytokines from muscle cells, all of which impair insulin action. We hypothesize that macrophages in both muscle and adipose tissues promote the development of metabolic syndrome during obesity. To test this hypothesis, adipose (Aim 1) and muscle tissue (Aim 2) from obese, insulin resistant compared to normal human subjects will be characterized. Macrophage number, the proportion of macrophages that are classically versus alternatively activated, and the proportion of macrophages in crown-like structures (in adipose) will be quantified, along with markers of the TGF2 pathway, fibrosis, hypoxia, tissue remodeling and vascularity. To specifically determine the role of TSP1 and the TGF2 pathway in insulin resistance, adipose and muscle tissue fibrosis, hypoxia, and macrophage function will be analyzed in high fat fed TSP1 null mice in Aim 3. A peptide that specifically blocks TSP1- mediated TGF2 activation will also be used in wild type, high fat fed mice to test the hypothesis that macrophage infiltration and fibrosis in adipose and muscle will be ameliorated in the absence of TSP1, resulting in improved insulin sensitivity. Aim 4 is the intervention aim in which the mechanisms underlying the beneficial effects of aerobic exercise training on insulin sensitivity will be determined. Following 12 weeks of training in obese and lean subjects, changes in muscle and adipose macrophage number, activation state, the TSP1/TGF2 pathway, and inflammation in response to a single bout of eccentric exercise will be quantified relative to insulin sensitivity. Finally, in Aim 5, the effects of different macrophage populations on adipocytes and muscle cells will be studied in vitro. The expression of genes involved in adipogenesis, extracellular matrix, and the TGF2 pathway will be monitored in adipocytes following macrophage co-culture. These will also be quantified in muscle cells, as well as the response to insulin. We will determine whether mechanical stimulation of myotubes to mimic exercise will alter the muscle cell response to macrophages.